INDEX

Explanations

keywords related to interrogations

oai_token-act-pair · gpt-3.5-turbo

terms related to interrogation and questioning processes

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Offline

-0.73

 spare

-0.67

minecraft

-0.66

ulz

-0.65

aim

-0.63

fortune

-0.63

cakes

-0.62

jri

-0.62

buy

-0.62

ensical

-0.62

POSITIVE LOGITS

 interrog

1.23

 interrogation

1.17

 interrogated

1.01

 Techniques

0.86

 techniques

0.84

 questioning

0.81

isen

0.77

 probing

0.76

 torture

0.76

 tactics

0.74

Activations Density 0.018%

keywords related to interrogations

terms related to interrogation and questioning processes

No Comments

No Known Activations